Mentor Support and Positive Youth Development in Urban Informality: Evidence from Secondary School Students in Kibra, Kenya ()
1. Introduction
Adolescents around the globe are faced with structural barriers that impede developmental processes such as unemployment, poor transition from schools to work, economic vulnerability, and involvement in violence and drug abuse [1] [2]. This has led to the emergence of the Positive Youth Development (PYD) paradigm as an approach that explains how youth can develop competencies such as confidence, connections, competence, character, caring, and contribution [3] [4]. Research conducted in diverse settings shows that PYD programs enhance socioemotional development, self-efficacy, and reduce risky behaviors, even though their effectiveness varies based on the design of the program, intensity, and the setting [5].
Mentor support has increasingly been identified as a key relational resource in adolescent development. Mentorship provides for systematic and non-familial associations that help provide young individuals with emotional, behavioral support, and social learning experiences [6]. According to previous literature, this result depends upon certain relational aspects, such as quality, duration, role modeling, and guidance and counseling [7]. Quality mentorship would facilitate trustworthiness, honesty, and transparency, whereas role modeling will assist with identity formation and value creation. Meanwhile, guidance and counseling will foster problem-solving skills and goal-setting abilities along with psychosocial adaptation. Despite the progress made in this field, however, research results continue to be inconsistent, often due to inconsistency in mentoring programs and the overall context in which these processes take place [6] [7].
Sub-Saharan Africa experiences this situation most strongly because of its fast-growing cities and increasing population and its existing social disparities and lack of employment options for young people. In 2023, about 53 million young people in the region were Not in Employment, Education, or Training (NEET), representing 21.9 per cent compared to the global rate of 20.4 per cent [8]. The region also continues to fall behind on Sustainable Development Goal target 8.6, which focuses on reducing NEET rates [9] [10].
These factors become evident in urban slums in Kenya, particularly Kibra, which is one of the biggest informal urban settlements in Nairobi and one of the biggest poor communities in Africa [11]. These conditions increase the risk of school dropout, weak skills development, and poor labour market readiness among young people. Although youth development and mentorship programmes operate in such communities, there is still limited empirical evidence on how specific dimensions of mentor support influence Positive Youth Development (PYD) outcomes in these settings. Available empirical research in Kenya tends to explore school-based guidance programs or general youth programs and neglects the relational aspect of mentoring, which is critical in an urban and high-risk setting like this one [12]. To fill this gap the gap, this study explores the extent to which relationship quality, duration of mentoring, role modeling, and guidance and counseling influence PYD among secondary school students in Kibra. The rest of the article is organized around the relevant literature, methodology, findings, and implications for policy and practice.
2. Literature Review
In this section, the existing literature that is pertinent to the topics of mentor support and Positive Youth Development (PYD) in urban informal settlements is explored. Theoretical perspectives, empirical studies, and conceptual models that inform the study are reviewed with a focus on the interaction of the dimensions of mentorship, quality of the relationship, duration, role modeling, positive affirmation, and guidance and counseling with the youth development outcomes. Moreover, the section elaborates on the influence of parental and contextual elements on mentorship programs’ effectiveness. This worldwide and local evidence gathering through the literature review indicates the gaps, challenges, and insights that affect the research design and framework of the study.
2.1. Theoretical Framework
The research applies the PYD approach and the conceptual model of youth mentoring proposed by Rhodes to examine the potential relationship between mentor support and youth development outcomes for high school students in Kibra, Kenya. The mentor support is assessed through the following five dimensions: quality of relationship, duration, modeling, positive affirmation, and guidance/counseling. The dimensions are perceived as interdependent processes that facilitate the ability of mentors to contribute to social, emotional, and cognitive development, with parental support being a moderating variable influencing the outcome of mentorship. The dual perspective portrays the development process in urban slums as a multi-dimensional construct based on the portrayal of youth development in such environments.
PYD provides the foundation for the primary views that center on the strengths and resources of the young people rather than on the problems. The key pillars are made of the six facets which are: competence, confidence, connection, character, caring and contribution [4] [13]. These areas are regarded as mutually reliant: a lack of one can adversely affect the whole development. Mentor support is seen as an external developmental asset that interacts with these internal strengths. Nonetheless, the practical application is constrained in informal settlements by a number of factors, including the use of measurement tools born in Western contexts that may not totally capture the local youth experiences [14].
Rhodes’ conceptual model coincides with PYD as it has already identified the relational processes through which mentorship may influence the young people. A high-quality, trust-based mentor-mentee relationship is thought to promote the development of social-emotional skills, identity making, and intellectual abilities [15] [16]. The aspects of mentorship, role-modeling, advising, and supporting, are presumed to be the means through which positive traits are developed by relational interactions. On the contrary, inconsistent, brief, or culturally misaligned mentoring may lead to minor advantages, or in rare cases, even more stress for the mentees, which indicates the latter scenario is worse than not having a mentor at all.
The two models still put a strong emphasis on the factors of the context. The prevailing poor socio-economic conditions in Kibra such as poverty, school disruptions, family tensions, and lack of community resources are driving force of the mentoring situation. Parents’ participation in mentoring children can obscure or enhance their impact, but still, the differences in support, engagement, and home atmosphere lead to different outcomes. Therefore, it is not to be overlooked that different young people may receive different treatment by the same mentorship programs and that a diversity of environmental and family factors has to be considered.
The integrated framework has also taken into consideration the methodological and conceptual limitations. The use of cross-sectional research designs makes it impossible to determine the causal relationship between mentor support and positive youth development outcomes. The different skill levels of the mentors, the length of the program and the degree of engagement of the mentees are factors that can lead to the quality of the effects produced being different. For instance, in the case of a mentor giving strong support or affirmation in one area, he or she might do the opposite by giving no support at all in the area of career orientation or social-emotional coaching, hence, the mentee’s development will be limited in terms of being holistic. The risk of programmes overrating the impact of mentorship in comparison to home or community influences increases without diligent monitoring and adaptation being done.
To sum up, the integration of PYD and Rhodes’ model not only allows but also entails a comprehensive understanding of mentoring in urban informal settlements where the strengths of the mentoring interventions are revealed and the relational processes are seen. Accordingly, the framework elucidates that the provision of mentor support is an influence on youth development that is multi-dimensional and context-parallel at the same time, the factors being the quality, length, role-modeling, affirmation, guidance, and parental interaction of the relationship. On the other hand, all relationships are dependent on the specific context, and consequently, the programme effects should not be assumed to be equal, nor guaranteed, thus, the necessity for an interpretation with caution and the design of the programme that is adaptive.
2.2. Empirical Review: Mentor Support and Positive Youth Development
Mentoring support is frequently referred to as social capital which the youth can access for their personal growth and playful aspects of life; however, the realization of the same capital benefits may differ. Social capital is defined as the resources of relationships which are mostly young people’s, especially during life difficulties, to help them attain their goals [17]. Positive youth development approaches emphasize that both internal and external assets should be invested in. Nevertheless, many programs still operate under the mistaken belief that mere participation will bring about the real benefits. Ross and Tolan [18] perceive the relationships in the center of youth development as the most important ones but it is very common that different interventions make it hard to find good mentors, especially in urban informal settlements where the structural hindrances and the competition for grassroots and time engagement limit the accessibility. This concern opens up the realization that much more stringent studies will have to be carried out to figure out how mentor support functions in different youth contexts which are the very distinct settings.
Mentorship quality comprising trust, emotional bonds, reciprocity, etc. is considered extremely important yet very difficult to standardize or measure at the same time. It is a common belief that the results of such relationships would be similar, i.e. the growth of competence, confidence, and connection, however, it might also be the case that they would be different and it would depend on the mentor’s ability and involvement. Nouri et al. [19] indicated that the quality of relationships is based on cultural congruence, effective communication, and mutual regard. However, in practice, researchers very rarely take into account the mentors who are unreliable, unqualified, or very busy, as their presence might downgrade the overall impact of the program. When working in urban slums, mentors may face difficulties like lack of resources or school strikes which may make it impossible for them to cultivate and use the high-quality engagement method expected from them and thus, they may produce uneven PYD benefits.
Mentor’s time and continuity in the relationship are considered crucial, but the evidence for this is open to interpretation. It is a common assumption that a long-term mentoring relationship results in a higher level of competence, self-esteem, and prosocial conduct [20]. Woodall et al. [21] however caution against the belief that short-term or prematurely interrupted relationships may yield little benefit or worse, instill stress in the youth. Still, the research generally does not consider the fact that informal settlements are in a state of continuous change, whether in terms of population mobility, family disruptions, or irregularities with the program delivery, all of which may negatively affect the duration effects. Thus, while duration matters, it is insufficient alone without attention to relational quality and consistent engagement.
One of the main characteristics of the supporter’s assistance is role modeling which is considered a significant feature. The behavior and decision-making of the mentors who are leading the positive are giving the mentees aspirational examples to follow and thus influencing their attitudes towards academic success and socially accepted ways of behaving. Mentors who are non-familial adults, as educators, program directors, or civil activists, can still support children in a non-peer and non-parent way [22]. However, the strong impact of model behavior is really a question of how much the mentors are seen, trusted, and their presence in the life of the child is consistent. Consequently, model behavior becomes extremely context-dependent and its impact is probably reduced if there are no constant and culturally appropriate interactions.
Confidence, motivation, and self-efficacy are all positively confirmed but the experience is not the same for all. The mentees who feel that they are being encouraged and accepted by the mentors declare a high level of competence, connection, and character development [23]. However, the reliance on self-reported perceptions constitutes a risk of bias, while on the other hand, inconsistent application may lead to some mentees feeling less supported. Batory-Ginda [24] endorses the assertion that affirmation empowers self-efficacy and goal orientation but, in the same breath, he points to the lack of practical ways to combine affirmation with guidance, role modeling, or mentoring longevity. In informal settling contexts where the youth are subjected to multiple stressors, the inconsistent affirmation might curtail the overall developmental potential of mentoring programs.
The provision of guidance and counselling forms the last important aspect of mentor support. The structured guidance in academic, career, and personal development corresponds to the Rhodes’ model of youth mentorship where the relationship between mentor and mentee is the main feature that leads to the growth and adjustment of the young person [25]. Mentoring in counseling and consulting has been documented and verified as one of the means through which mentor support can provide emotional or behavioral support [26]. Nevertheless, the situation has been such that some mentors are exceedingly supportive while others are hardly supportive or even not involved at all in the counseling process by giving advice or checking on the mentee. This reality makes it very challenging to detect the influence of PYD and suggests that guidance interventions require clear standards, monitoring, and contextual flexibility for stronger developmental impact.
The accumulated evidence indicates that mentor support in urban informal settlements is multifaceted with different influences being dependent on the quality of the relationship, the length of time spent together, role modeling, affirmation, and guidance. There is a consistent connection of some dimensions, particularly structured guidance and sustained role modeling, to PYD indicators; however, there are also limitations such as gaps in implementation, contextual constraints, and lack of consistency in the engagement of mentors which are restricting the overall effectiveness. The combination of the quantitative and qualitative insights reveals that it is not enough to provide mentors, but also that they should be trained, committed, and sensitive to the needs of the youth in their particular social and environmental contexts.
2.3. Conceptual Framework for Mentor Support and Positive Youth Development
Figure 1 below outlines the conceptual framework used in the study.
Source: Conceptualized by author (2024).
Figure 1. Conceptual framework.
In this framework, mentor support forms the core focus of Positive Youth Development (PYD) among secondary school students in Kibra. It is examined through five connected dimensions: quality of relationship, duration, modelling, positive affirmation, and guidance/counselling. These dimensions operate together within the mentoring process and describe how students experience support in school settings. PYD is treated as the outcome and is captured through six areas: competence, confidence, connection, character, caring, and contribution, consistent with PYD scholarship that views development as a set of interlinked strengths rather than a single attribute [14].
The framework also draws on Rhodes’ mentoring model, which explains mentoring as a relational process shaped by trust, emotional connection, identity formation, and social learning. Within this view, variation in the quality and consistency of mentoring is associated with differences in how these processes are experienced [16]. In this study, mentor support is positioned as an external developmental asset that operates through these relational pathways rather than as a stand-alone factor.
Parental support is factored in within the broader framework and is not examined as a predictor but as a background context. It is defined by the differences in the contexts of families, like economic pressure and family stability, and how these relate with adolescents’ view and their interactions with mentoring relationships. In the case of Kibra, these contextual variables coexist with the mentoring programs within schools and contribute to the context where mentoring occurs.
In general, the framework ensures that mentoring processes remain central while recognizing that mentoring experiences vary among students because of the social context around them and their relational involvement during mentoring interactions.
3. Methodology
A convergent parallel mixed-methods design was used in this study to research the relationship between mentor support and Positive Youth Development of adolescents in Kibra Sub-County, Nairobi. The design involved simultaneous data collection of survey and focus group data. Surveys revealed the relationships between the support from mentors and the outcomes of the young people. Focus groups explored how young people were mentored in the context of their daily lives. By portraying figures with human voices, the dual approach secured greater trustworthiness. Moreover, it laid out a broader spectrum of how mentorship is hooked up to six PYD domains which include competence, confidence, connection, character, caring, and contribution [4].
For the quantitative study, the sample was made up of 3560 secondary-school adolescents who were participating in ten school-based mentorship programs across Kibra. Ten research areas were selected systematically, and study respondents recruited based on the program size. A total of 359 youths were targeted as the sample with a five percent margin of error. The resulting dataset was made up of 349 program participants and 102 non-participants, with a total response rate of 94 percent. Participants included students from Form 1 to Form 4 who had been in the programs for a minimum of one year. A written assent was given by each of the enrolled participants. Parents or guardians, in accordance with the Children Act 2022, gave an informed signed consent. On-participants were drawn from church youth groups, which were selected for convenience of access and familiarity with the community.
Mentor support was assessed through the application of adjusted items from the Mentor Support Provisions Scale. The support which was detected, the duration of the mentoring relations, role modeling, teaching, and counseling were all focused on in the questioning process. The five-point scale from totally disagree to totally agree was used by the survey participants as a way to express their opinion on the statements [27]. The scale verified its internal consistency through a high Cronbach’s alpha coefficient of 0.91. Participation has an alpha of 0.89 and self-esteem 0.87. On the other hand, parental support was associated with alpha of 0.918 measured by the 8-item scale while community engagement had an alpha of 0.701 after being assessed through 7 items. Based on the Lerner’s Six Cs framework, the PYD outcome was the reliability scores of 0.80 - 0.85 [28]. The vocabulary, administration methods, and field logistics were modified to fit the local language and context.
The quantitative data statistical analysis was done using both descriptive and inferential analysis methods. Frequencies, means, and standard deviations were employed in order to describe the key variables. The PYD concept was formulated by creating an aggregate index from the six dimensions of PYD. Competence, Confidence, Connection, Character, Caring, and Contribution constituted the PYD constructs which were all evaluated via Likert scale questions. The individual responses to each construct question were aggregated together to create an overall score of the respective dimension. The scores were equally weighted to create an overall score for PYD for each respondent. The resulting score was treated as continuous in regression analysis and also grouped into three equal-interval levels for comparison purposes.
To examine associations between mentorship, family influences, and Positive Youth Development (PYD), cross-tabulations and chi-square tests were used after PYD was grouped into three ordered categories: low, moderate, and high (3 - 5, 6 - 8, and 9 - 10). To assess how mentorship dimensions relate to overall PYD when considered together, linear regression was applied with PYD treated as a continuous outcome. This procedure is well suited for the nature of the data as PYD was created using index construction; thus, variation among individuals can be incorporated into the study. Prior to model estimation, the important assumptions of linear regression such as linearity, independence of errors, homoscedasticity, and multicollinearity were tested [29]. All statistical tests were done in SPSS, using α = 0.05.
The qualitative information was collected using 15 focus group discussions and 10 key informant interviews conducted with the program managers. Focus group discussion consisted of both mixed gender discussions as well as single gender discussion groups to ensure participation from all members. Notes for all discussions were taken down using handwritten notes in the language spoken in the locality and subsequently crosschecked by another researcher to minimize transcription errors. Analysis was done using the thematic approach proposed by Braun and Clarke [30] and included multiple readings of the data, coding, and generation of themes.
The data integration process used a joint-display approach, where qualitative themes were aligned alongside quantitative results to support structured comparison across datasets [31]. This allowed the two forms of data to be viewed in relation to the same analytical focus without merging them into a single dataset. The integration focused on examining areas of convergence and divergence between the qualitative and quantitative strands, particularly around mentorship experiences and PYD as conceptualized in the study framework.
Ethical approval was granted by the St. Paul’s Ethics Review Committee and the National Commission for Science, Technology and Innovation (NACOSTI). Pseudonyms for all respondents as well as the provision of secure data storage ensured anonymity and confidentiality. Every communication took place through research assistants who were well versed in child protection protocols. Participation was verified through attendance logs and mentor records, which also helped to reduce reporting errors and thus reinforced the credibility of the data. The convergent parallel design was not just theoretical but actually served a practical purpose. Theoretically, it allowed measuring the impact of the mentor support on youth development while also capturing the daily processes that were behind that impact.
Overall, the mixed approach led to a thorough and reliable understanding of the functioning of mentorship in the Kibra slum area. The positivity and negativity of the interventions were not only distinguished but also described, which is in line with the current literature on the subject of the same kind of mentoring in low-income urban areas [32] [33].
4. Study Findings and Discussion
The purpose of this study was to explore the links between mentoring support for young people and positive youth development (PYD) in a sample of secondary school adolescents in Kibra Sub-City of Nairobi, Kenya (an urban informal settlement). This study used a mixed-methods approach using survey data (quantitative), focus groups (qualitative), key informants (qualitative), and individual interviews (qualitative). Analysis included descriptive statistics, cross-tabulations and linear regression (quantitative), as well as thematic analysis on qualitative data. Mentor support was examined across five dimensions; mentor relationship, mentor longevity, mentor (role modelling), positive affirmation and overall guidance. Overall, adolescents reported predominantly positive experiences with regards to mentoring in the domains of helping to find assistance in their academic work, aiding decision-making and providing emotional support. The report contains tables of findings for presentation and ease of reference.
Table 1 reflects favorable mentor connections, marked by high levels of enjoyment of mentoring (M = 4.47, SD = 0.65) and supportive expectations (M = 4.49, SD = 0.64). This suggests that relational engagement in the mentor-mentee connection is strong. Nevertheless, low scores on time management help (M = 3.08, SD = 1.28) and dealing with academic disappointments (M = 3.45, SD = 1.27) suggest inconsistent application of structured academic assistance amidst good relational connections.
Another significant measure examined was mentorship longevity, presented in Table 2. The findings show variation in duration of exposure across classes in school-based mentoring programs. Form One and Form Two students reported an average of 3 years of mentorship (Mean = 3.0, SD = 2.9 and 3.0). Form Three students reported an average of 5 years (Mean = 5.0, SD = 4.6), while Form Four students reported the highest average at 6 years (Mean = 6.0, SD = 4.3). Overall, the mean duration of mentorship was 4 years (SD = 3.8), with a range from 1 to 16 years. The wide range suggests uneven exposure to mentorship across students, with some reporting long-term engagement and others relatively recent entry into the program.
Table 3 presents the descriptive statistics regarding the role modelling. It revealed that through role modelling one of the major subcategories of mentor support was identified in the mentorship programme. The findings reveal that role modeling is indicated in the perceptions of mentoring received by the participants
Table 1. Descriptive statistics on mentor relationship.
Statement |
SD [f, (%)] |
D [f, (%)] |
N [f, (%)] |
A [f, (%)] |
SA [f, (%)] |
Mean |
Std. Dev |
I enjoy the experiences I derive from my mentor. |
2 (0.6) |
3 (0.9) |
13 (3.7) |
142 (40.7) |
189 (54.2) |
4.47 |
0.65 |
I regularly apply skills learnt from my mentor. |
14 (4.0) |
34 (9.7) |
46 (13.2) |
158 (45.3) |
97 (27.8) |
3.83 |
1.05 |
I feel overwhelmed by mentors’ expectations. |
3 (0.9) |
3 (0.9) |
8 (2.3) |
146 (41.8) |
189 (54.2) |
4.49 |
0.64 |
My mentor has taught me to plan and use time optimally. |
54 (15.5) |
72 (20.6) |
62 (17.8) |
115 (33.0) |
46 (13.2) |
3.08 |
1.28 |
When my academic career goals were not met, I receive assistance from my mentor to develop better coping strategies. |
36 (10.3) |
56 (16.0) |
41 (11.7) |
147 (42.1) |
69 (19.8) |
3.45 |
1.27 |
I have been motivated to make well-informed personal decisions about my academic and professional ambitions. |
22 (6.3) |
30 (8.6) |
29 (8.3) |
139 (39.8) |
129 (37.0) |
3.92 |
1.15 |
My mentor has followed up on my plans to develop better academic skills by asking me about my actual progress. |
15 (4.3) |
33 (9.5) |
36 (10.3) |
147 (42.1) |
118 (33.8) |
3.92 |
1.08 |
My mentor helps me explore realistic options and provides guidance on how to attain my academic objectives. |
5 (1.4) |
8 (2.3) |
18 (5.2) |
147 (42.1) |
171 (49.0) |
4.35 |
0.77 |
Table 2. Mentorship longevity.
Current class |
Mean |
Std. deviation |
Minimum |
Maximum |
Form one |
3 |
2.9 |
1 |
13 |
Form two |
3 |
3 |
1 |
14 |
Form three |
5 |
4.6 |
1 |
16 |
Form four |
6 |
4.3 |
1 |
16 |
Total |
4 |
3.8 |
1 |
16 |
Table 3. Descriptive statistics on role modelling.
Statement |
SD [f, (%)] |
D [f, (%)] |
N [f, (%)] |
A [f, (%)] |
SA [f, (%)] |
Mean |
Std. Dev. |
I consider my mentor my role model. |
9 (2.6) |
21 (6.0) |
22 (6.3) |
128 (36.7) |
169 (48.4) |
4.29 |
0.99 |
I admire my mentor and want to be successful like him/her. |
7 (2.0) |
13 (3.7) |
26 (7.4) |
125 (35.8) |
178 (51.0) |
4.3 |
0.95 |
My mentor inspires my confidence about the future. |
3 (0.9) |
14 (4.0) |
25 (7.2) |
137 (39.3) |
170 (48.7) |
4.34 |
0.87 |
My mentor and I share the same values. |
4 (1.1) |
4 (1.1) |
8 (2.3) |
156 (44.7) |
177 (50.7) |
4.44 |
0.74 |
My mentor inspires me because he/she always gives back to the community. |
19 (5.4) |
58 (16.6) |
57 (16.3) |
133 (38.1) |
82 (23.5) |
3.69 |
1.15 |
My mentor has helped nurture my emotional and social skills. |
13 (3.7) |
23 (6.6) |
36 (10.3) |
145 (41.5) |
132 (37.8) |
4 |
0.97 |
My mentor makes me feel comfortable enough in our relationship that I can share openly about sensitive topics. |
19 (5.4) |
34 (9.7) |
40 (11.5) |
142 (40.7) |
114 (32.7) |
3.91 |
1.05 |
My mentor is my primary guide in all my career and professional decisions. |
5 (1.4) |
9 (2.6) |
8 (2.3) |
158 (45.3) |
169 (48.4) |
4.38 |
0.75 |
My mentor accords practical approaches whenever I have a problem by sharing his/her personal experiences. |
10 (2.9) |
9 (2.6) |
21 (6.0) |
151 (43.3) |
158 (45.3) |
4.28 |
0.82 |
from the program in terms of the highest means being recorded in value alignment (M = 4.44, SD = 0.74), career guidance (M = 4.38, SD = 0.75), and future optimism (M = 4.34, SD = 0.87). This suggests that there is high agreement among participants regarding these areas of mentoring support. The rest of the items also show mostly positive responses from the learners. They show high agreement that mentors are role models (M = 4.29, SD = 0.99) and mentors give advice and guidance based on their own experiences (M = 4.28, SD = 0.82). Regarding emotional and social support, moderate agreement was found among the respondents (M = 4.00, SD = 0.97), and slightly low scores were obtained for openness to discuss personal issues (M = 3.91, SD = 1.05).
Building on the patterns of mentorship experience, results in Table 4 reveal that there is relatively high positive affirmation across all items, where scores range from 4.11 to 4.54 on average.
Table 4. Descriptive statistics on positive affirmation.
Statement |
SD [f, (%)] |
D [f, (%)] |
N [f, (%)] |
A [f, (%)] |
SA [f, (%)] |
Mean |
SD |
My mentor continuously appreciates the effort that I make in my academics. |
9 (2.6) |
7 (2.0) |
16 (4.6) |
140 (40.1) |
177 (50.7) |
4.34 |
0.83 |
My mentor believes in me. |
10 (2.9) |
6 (1.7) |
27 (7.7) |
131 (37.5) |
175 (50.1) |
4.31 |
0.86 |
When my school grades are not good, my mentor still encourages me to be focused. |
3 (0.9) |
4 (1.1) |
9 (2.6) |
121 (34.7) |
212 (60.7) |
4.54 |
0.72 |
When my school grades are not good, my mentor helps me determine the best solutions. |
8 (2.3) |
3 (0.9) |
8 (2.3) |
153 (43.8) |
177 (50.7) |
4.4 |
0.82 |
My mentor challenges me towards my purpose. |
14 (4.0) |
12 (3.4) |
23 (6.6) |
166 (47.6) |
134 (38.4) |
4.13 |
0.98 |
My mentor recognizes and affirms all efforts towards my career development. |
12 (3.4) |
7 (2.0) |
36 (10.3) |
157 (45.0) |
137 (39.3) |
4.15 |
0.92 |
My mentor has helped me identify my strengths and guides me on how to focus on them. |
8 (2.3) |
18 (5.2) |
40 (11.5) |
143 (41.0) |
140 (40.1) |
4.11 |
0.95 |
From the study, about 90% of students give encouraging ratings on the encouragement they receive from their mentors. The encouragement received after poor academic performance has the highest score (M = 4.54, SD = 0.72) compared to that used in solving the academic problems of the learner (M = 4.40, SD = 0.82). The other areas that may be considered as academic encouragement include encouragement received because of effort put forth when learning (M = 4.34, SD = 0.83), as well as the belief and encouragement in learners’ capabilities (M = 4.31, SD = 0.86). The other areas have fairly positive ratings but with lower scores. These include encouragement received as a result of effort put forth in terms of career building (M = 4.15, SD = 0.92), as well as orientation to personal destiny (M = 4.13, SD = 0.98).
Overall, the findings indicate a positive perspective on the school-based mentorship program in schools across all five categories of mentor support from the point of view of the students at Kibra. Engagement in terms of the mentor relationship dimension shows very positive perceptions from the learners regarding their experiences with the mentors in addition to supportive interaction, except for time management and dealing with any academic frustrations, which vary in terms of response. Characteristics of role modelling involve strong consensus on common values, career path, and orientation towards the future. In contrast, openness to discussion of personal issues and community-oriented influence is not as consistent across all respondents. Similarly, positive affirmation is also strong, particularly in the manner that the mentors react when facing academic problems, motivate them to put in more efforts, and help build confidence among the learners. In terms of overall descriptive patterns, there seem to be differences in experiences related to particular mentorship components.
Following this, the analysis sought to establish any correlations between mentorship dimensions and Positive Youth Development (PYD) levels. The data was first organized in cross-tabulation tables that demonstrated the distribution of participants among the three mentorship dimensions in regard to their PYD levels – low, moderate, and high. Then, chi-square (χ2) tests were conducted to confirm the significance of differences observed between groups. This method was chosen because the two variables (mentorship dimensions and PYD) had been dichotomized; hence, the comparison between them was based on the positions they occupied relative to one another.
From the findings, there is no any statistically significant relationship between mentor relationships or role models and youth development (mentor relationship: χ2 = 2.17, p = 0.338; role modelling: χ2 = 3.65, p = 0.161). There is also a nearly equal distribution of responses across the three categories of development in relation to both mentor relationships and role models, which cannot be interpreted as having any statistical significance. However, there is a statistically significant relationship between the mentor’s positive affirmation and youth development (χ2 = 9.05, p = 0.011), with almost all youths responding in the highest category of youth development. A similar pattern exists for mentor support (χ2 = 6.45, p = 0.011), with the vast majority of responses occurring in the highest category of youth development. The data thus suggests that youth development has a greater association with mentor support and encouragement than with the relationships and role modelling aspects of mentorship. Table 5 summarizes the findings.
Table 5. Cross-tabulation of Mentorship Dimensions and Positive Youth Development (PYD.
Mentorship dimension |
% with high PYD (Low) |
% with high PYD (Medium) |
% with high
PYD (High) |
χ2 |
df |
p-value |
Mentor relationship |
100% |
92% |
96% |
2.17 |
2 |
0.338 |
Role modelling |
100% |
89% |
96% |
3.65 |
2 |
0.161 |
Positive affirmation |
100% |
84% |
97% |
9.05 |
2 |
0.011 |
Mentor support |
– |
84% |
96% |
6.45 |
1 |
0.011 |
The chi-square analyses demonstrated differences between groups, however, they did not indicate how different mentorship factors interact. To determine how the mentorship factors were related to one another, a linear regression model was created with the PYD represented as a continuous composite (or total) score on the mentor measure. This method was appropriate because the PYD was constructed as an index; although it was later grouped into 4 categories for the purposes of comparison. All the requisite model diagnostics were performed prior to interpreting the results of the regression model. Linear relationships were assessed via plots of the standardized residuals by the predicted value with no evidence of significant deviation.
Durbin-Watson statistics confirmed independence among the errors and fell within acceptable limits. Residual plots revealed that there was no major fluctuation in the distribution of the stripped, indicating the presence of homoscedasticity. Path VIF’s were calculated to determine multicollinearity to determine how closely the different mentorship factors were correlated with one another. All of the calculated VIF values were within acceptable thresholds. Thus, these tests all appear to support the use of the regression model for this analysis. The regression results are presented in Table 6.
Table 6. Regression analysis on mentor support and positive youth development.
Coefficients |
B |
Std. Error |
Beta |
t |
Sig. |
(Constant) |
1.966 |
0.183 |
– |
10.739 |
0.000 |
Mentor relationship |
0.078 |
0.029 |
0.092 |
2.646 |
0.001 |
Longevity |
0.004 |
0.002 |
0.11 |
2.446 |
0.015 |
Role modelling |
0.174 |
0.049 |
0.229 |
3.563 |
0.000 |
Positive affirmation |
0.016 |
0.007 |
0.021 |
2.299 |
0.025 |
Model summary: R = 0.561; R2 = 0.315; Adjusted R2 = 0.304; Std. Error of the estimate = 0.352.
From Table 6, the regression model is statistically significant (F = 31.295, p < 0.001) and explains about 31.5% of the variability in PYD (R2 = 0.315). The regression coefficient for role modeling is the highest (B = 0.174, p < 0.001). Mentor relationship also has a positive and statistically significant coefficient (B = 0.078, p = 0.001). The coefficient for longevity is smaller (B = 0.004, p = 0.015), suggesting a small difference in PYD between the different levels of exposure.
Positive affirmation has a statistically significant coefficient (B = 0.016, p = 0.025), which is smaller than that of the other factors. This is in contrast with the earlier chi-square results, which show greater associations between the factor and PYD categories. Based on these results, it may be concluded that, in the context of the full model, role modeling and mentor relationship have stronger associations with PYD compared with the other factors.
Overall, the quantitative measurements indicate mostly positive perceptions regarding mentorship across all dimensions, in which high perceptions are mostly provided on relational involvement, role modeling, and positive reinforcement. On other dimensions, particularly those that deal with the structured learning process, varying perceptions can be observed among the respondents. Based on the analysis done through cross-tabulation and chi-square testing, it is observed that there is no relationship between the PYD categories and mentorship dimensions. However, positive affirmation and mentor support exhibit significant associations, whereby more responses are found in the high PYD category. By contrast, the regression findings bring another angle to the assessment of associations between mentorship and PYD, where all dimensions of mentorship are considered simultaneously. Here, however, each dimension makes unique contributions to PYD in different amounts. For example, role modeling and mentor relationship have large coefficients compared to longevity and positive affirmation even though they are statistically significant.
Taken together, the results suggest that PYD is more consistently aligned with the functional and interactional aspects of mentorship (such as affirmation and structured support) than with relational proximity alone. The lack of significance in some chi-square comparisons, alongside significant regression coefficients, points to the influence of model specification: group-based comparisons may mask variation that becomes visible when PYD is treated as a continuous construct. At the same time, the smaller coefficients for some variables suggest that not all mentorship dimensions carry equal analytical weight when considered simultaneously.
The qualitative data provide contextual depth to these patterns. Some of the interviewees mentioned that the application of positive affirmations has increased their confidence and has managed to change the way they see things. A participant responded:
“My mentor encourages and talks to me positively for most of the time. We set the academic, future career and relational goals every holiday when we meet. She gives me positive feedback ... She always says, ‘Getty, you have a God-given destiny… you can do it… you are more than able with the help of the Lord.’ She prays with me and for me” (FM #7, FGD 1-1, Project #00A).
Another student highlighted how affirmation was coupled with structured guidance, noting,
“My mentor is a Muzungu (White Man) Missionary. Together we have discussed my strengths and weaknesses and what I can do to avoid conflict. Then you plan how to work together, the academic, career and life goals” (MR #3, FGD 1-1, Project #00I).
In the PYD framework, youth outcomes are understood through how personal capabilities interact with external support systems, with mentorship forming a key part of that support. Existing studies show that this support is not always consistent in practice. Ross and Tolan [18] note that poorly prepared mentors tend to limit the development of confidence, autonomy, and social attachment. Deane et al. [20] find that there is a correlation between long-term mentoring and greater levels of self-respect, self-efficacy, and social skills; however, the benefits diminish once interruptions occur. Similarly, Astrove et al. [23] demonstrate a link between positive reinforcement in mentoring and self-concept and goal orientation but unequal access produces inconsistency across individuals.
In line with such research findings, the current study suggests that mentorship seems to be significant not just by itself, but in terms of its delivery, maintenance, and consistency. Therefore, PYD characteristics are better related to patterns of guidance, relationship continuity, and reinforcement. These elements appear more central than the presence of a mentor on its own. Parental support is not included as a statistical predictor in the model. It appears only in qualitative accounts and supporting literature to help explain how adolescents describe their lived environments. It provides context rather than an interacting or combined effect with mentorship, and interpretation of PYD remains based on the measured mentorship dimensions alone.
The above findings put across this notion that mentoring works but depends on how it is implemented. These new findings fall within the above general context. It would appear that mentoring can operate by way of networking and mentoring systems but not by sheer presence only. This implies that those programs which base their interventions on mentoring without putting into consideration consistency and many other such things would fail to deliver anything consistently.
5. Conclusions
The study aimed at understanding how mentorship played a part in the Positive Youth Development (PYD) process of secondary-school adolescents in Kibra, Nairobi’s informal settlement. The results reveal that mentoring contributes to the youths’ self-confidence, resilience, motivation and better focus on their goals. Mentored youths are even more assertive in their claims that they are equipped with fighting the hardship, mingling with the right crowd, and taking the lead. Mentoring enables the youth to have non-academic lives with the help of emotional stability, social skills, and general well-being as the support provided. The mentor is the one who gives this support by being there, offering encouragement and showing the way. Under the care of the mentor, the youth tend to be more immune to the bad and at the same time, they receive support from a holistic source in their winning the battle for being good citizens. The research recommends that if the youth’s potential is to be fully revealed and utilized, this would come about through the well-planned and continuous mentoring programs.
The study contributes to new knowledge. Role modeling was identified as the major indicators of personal growth, whereas the relationship factors including the affirmation from the mentor and the perceived quality of the relationship were rated very high descriptively, yet label almost no unique predictive power. This is an instance where perceived support does not engage with measurable developmental outputs, thus stipulating the individual and environmental factors as moderators. Moreover, the study implies mentor backing not just for the individual but also the development of the youth, who by their presence contribute to schools and communities, thereby highlighting the larger social benefit of structured out-of-school mentorship programs. The disclosed understandings extend the already established knowledge about the impact of different mentorship techniques on the promotion of PYD among youths in urban informal settlements, as per their effectiveness or otherwise.
The study had multiple limitations. First, the very high relationship quality and positive affirmation of the mentor could only account for a small part of the variance in the developmental outcomes, which indicates that the support from family, peers, or teachers, which were not included in this research, might be moderating the effects of mentorship. Secondly, there are ceiling effects implied by the high consensus on various items, which in turn, reduce the power to detect nuanced differences among subgroups. Thirdly, the study being context-specific, particularly emphasizing secondary-school adolescents in Kibra, presents a significant limitation for the extensiveness of the findings to other socioeconomic or cultural contexts. Future research will need to analyze the interplay between mentorship and broader support systems, to try out the mentoring models in various settings and to explore the ways for the enhancement of the overall mentoring, like developing social and emotional skills or engaging the community.